Covid-19 Analysis to visualize and understand the pandemic cases using different Techniques.

Coronavirus disease 2019 (COVID-19) time series lists confirmed cases, reported deaths, active cases and comparison with other Epidemics. Data are disaggregated by country (and sometimes subregion). Coronavirus disease (COVID-19) is caused by the Severe acute respiratory syndrome Coronavirus 2 (SARS-CoV-2) and has had a worldwide effect.

This notebook uses data from various sources to understand, analyze and visualize the changes in the number of cases through different visualization techniques and plots.

This dataset includes time-series data tracking the number of people affected by COVID-19 worldwide, including:

• confirmed tested cases of Coronavirus infection
• number of people who have reportedly died due to Coronavirus
• number of people who have reportedly recovered from it

Note: The Data collection for Recovered cases isn't quite accurate and has been stopped by a lot of countries and it is also found discrepancies in data if taken from multiple sources. Also a lot of recovery cases aren't reported and it is not possible to analyze them accurately. Yet the data reported till July 2020, is quite accurate to understand. You can select from Jan2020 to Jan 2021 to clearly see the transition in all the graphs.

Therefore the following two scenarios should also be kept in mind:

• Active Cases are calculated based on recoveries so this will not be also correct
• Anything that requires recoveries in the calculation isn't accurate, like deaths/100recoveries, etc.

Additional References for the above point:

• https://www.dallasnews.com/news/public-health/2020/05/19/why-arent-coronavirus-recoveries-always-reported/
• https://abc11.com/nc-coronavirus-recovery-cases-update/6127051/

Most of these reports are based on the US but mostly it is true for the rest of the world.

pip install folium

Requirement already satisfied: folium in c:\users\parth\anaconda3\lib\site-packages (0.12.1.post1)
Requirement already satisfied: branca>=0.3.0 in c:\users\parth\anaconda3\lib\site-packages (from folium) (0.4.2)
Requirement already satisfied: jinja2>=2.9 in c:\users\parth\anaconda3\lib\site-packages (from folium) (2.11.3)
Requirement already satisfied: requests in c:\users\parth\anaconda3\lib\site-packages (from folium) (2.26.0)
Requirement already satisfied: numpy in c:\users\parth\anaconda3\lib\site-packages (from folium) (1.20.3)
Requirement already satisfied: MarkupSafe>=0.23 in c:\users\parth\anaconda3\lib\site-packages (from jinja2>=2.9->folium) (1.1.1)
Requirement already satisfied: certifi>=2017.4.17 in c:\users\parth\anaconda3\lib\site-packages (from requests->folium) (2021.10.8)
Requirement already satisfied: urllib3<1.27,>=1.21.1 in c:\users\parth\anaconda3\lib\site-packages (from requests->folium) (1.26.7)
Requirement already satisfied: charset-normalizer~=2.0.0 in c:\users\parth\anaconda3\lib\site-packages (from requests->folium) (2.0.4)
Requirement already satisfied: idna<4,>=2.5 in c:\users\parth\anaconda3\lib\site-packages (from requests->folium) (3.2)
Note: you may need to restart the kernel to use updated packages.

pip install plotly

Requirement already satisfied: plotly in c:\users\parth\anaconda3\lib\site-packages (5.5.0)
Requirement already satisfied: six in c:\users\parth\anaconda3\lib\site-packages (from plotly) (1.16.0)
Requirement already satisfied: tenacity>=6.2.0 in c:\users\parth\anaconda3\lib\site-packages (from plotly) (8.0.1)
Note: you may need to restart the kernel to use updated packages.

#required packages and libraries

import plotly.express as px
import plotly.graph_objects as go
import plotly.figure_factory as ff
from plotly.subplots import make_subplots

import folium

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt

%matplotlib inline

import math
import random
from datetime import timedelta

import warnings
warnings.filterwarnings('ignore')

# color pallette
cnf ='#393e46'
dth ='#ff2e63'
rec ='#21bf73'
act ='#fe9801'

Dataset Preparation

import plotly as py
py.offline.init_notebook_mode(connected = True)

import os

try:
    os.system("rm -rf Covid-19-Preprocessed-Dataset")
except:
    print('File does not exist')

# cloning the data 
!git clone https://github.com/laxmimerit/Covid-19-Preprocessed-Dataset.git

fatal: destination path 'Covid-19-Preprocessed-Dataset' already exists and is not an empty directory.

# Understanding the Data
df = pd.read_csv('Covid-19-Preprocessed-Dataset/preprocessed/covid_19_data_cleaned.csv', parse_dates=['Date'])
df

	Date	Province/State	Country	Lat	Long	Confirmed	Recovered	Deaths	Active
0	2020-01-22	NaN	Afghanistan	33.93911	67.709953	0	0	0	0
1	2020-01-23	NaN	Afghanistan	33.93911	67.709953	0	0	0	0
2	2020-01-24	NaN	Afghanistan	33.93911	67.709953	0	0	0	0
3	2020-01-25	NaN	Afghanistan	33.93911	67.709953	0	0	0	0
4	2020-01-26	NaN	Afghanistan	33.93911	67.709953	0	0	0	0
...	...	...	...	...	...	...	...	...	...
212493	2022-01-29	NaN	Timor-Leste	-8.87420	125.727500	0	0	0	0
212494	2022-01-30	NaN	Timor-Leste	-8.87420	125.727500	0	0	0	0
212495	2022-01-31	NaN	Timor-Leste	-8.87420	125.727500	0	0	0	0
212496	2022-02-01	NaN	Timor-Leste	-8.87420	125.727500	0	0	0	0
212497	2022-02-02	NaN	Timor-Leste	-8.87420	125.727500	0	0	0	0

212498 rows × 9 columns

# We need to handle the NaN values which are there under the Province/State column
df['Province/State'] = df['Province/State'].fillna("")
df

	Date	Province/State	Country	Lat	Long	Confirmed	Recovered	Deaths	Active
0	2020-01-22		Afghanistan	33.93911	67.709953	0	0	0	0
1	2020-01-23		Afghanistan	33.93911	67.709953	0	0	0	0
2	2020-01-24		Afghanistan	33.93911	67.709953	0	0	0	0
3	2020-01-25		Afghanistan	33.93911	67.709953	0	0	0	0
4	2020-01-26		Afghanistan	33.93911	67.709953	0	0	0	0
...	...	...	...	...	...	...	...	...	...
212493	2022-01-29		Timor-Leste	-8.87420	125.727500	0	0	0	0
212494	2022-01-30		Timor-Leste	-8.87420	125.727500	0	0	0	0
212495	2022-01-31		Timor-Leste	-8.87420	125.727500	0	0	0	0
212496	2022-02-01		Timor-Leste	-8.87420	125.727500	0	0	0	0
212497	2022-02-02		Timor-Leste	-8.87420	125.727500	0	0	0	0

212498 rows × 9 columns

#importing the country_daywise, countrywise, daywise dataset
country_daywise = pd.read_csv('Covid-19-Preprocessed-Dataset/preprocessed/country_daywise.csv', parse_dates=['Date'])
countrywise = pd.read_csv('Covid-19-Preprocessed-Dataset/preprocessed/countrywise.csv')
daywise = pd.read_csv('Covid-19-Preprocessed-Dataset/preprocessed/daywise.csv', parse_dates=['Date'])

country_daywise

	Date	Country	Confirmed	Deaths	Recovered	Active	New Cases	New Recovered	New Deaths
0	2020-01-23	Afghanistan	0	0	0	0	0	0	0
1	2020-01-24	Afghanistan	0	0	0	0	0	0	0
2	2020-01-25	Afghanistan	0	0	0	0	0	0	0
3	2020-01-26	Afghanistan	0	0	0	0	0	0	0
4	2020-01-27	Afghanistan	0	0	0	0	0	0	0
...	...	...	...	...	...	...	...	...	...
145035	2022-01-27	Zimbabwe	229096	5324	0	223772	153	0	3
145036	2022-01-28	Zimbabwe	229333	5333	0	224000	237	0	9
145037	2022-01-29	Zimbabwe	229415	5333	0	224082	82	0	0
145038	2022-01-30	Zimbabwe	229460	5337	0	224123	45	0	4
145039	2022-01-31	Zimbabwe	229666	5338	0	224328	206	0	1

145040 rows × 9 columns

Worldwide Total Confirmed, Recovered Cases and Deaths

#looking at the number of confirmed cases
confirmed = df.groupby('Date').sum()['Confirmed'].reset_index()
confirmed

	Date	Confirmed
0	2020-01-22	557
1	2020-01-23	655
2	2020-01-24	941
3	2020-01-25	1434
4	2020-01-26	2118
...	...	...
738	2022-01-29	372549524
739	2022-01-30	374778211
740	2022-01-31	378391869
741	2022-02-01	381683860
742	2022-02-02	384701043

743 rows × 2 columns

#looking at the recovered cases
recovered = df.groupby('Date').sum()['Recovered'].reset_index()
recovered

	Date	Recovered
0	2020-01-22	30
1	2020-01-23	32
2	2020-01-24	39
3	2020-01-25	42
4	2020-01-26	56
...	...	...
738	2022-01-29	0
739	2022-01-30	0
740	2022-01-31	0
741	2022-02-01	0
742	2022-02-02	0

743 rows × 2 columns

#looking at the number of death cases
deaths = df.groupby('Date').sum()['Deaths'].reset_index()
deaths

	Date	Deaths
0	2020-01-22	17
1	2020-01-23	18
2	2020-01-24	26
3	2020-01-25	42
4	2020-01-26	56
...	...	...
738	2022-01-29	5659009
739	2022-01-30	5664564
740	2022-01-31	5674374
741	2022-02-01	5688629
742	2022-02-02	5699993

743 rows × 2 columns

#check for any null values
df.isnull().sum()

Date              0
Province/State    0
Country           0
Lat               0
Long              0
Confirmed         0
Recovered         0
Deaths            0
Active            0
dtype: int64

df.query('Country == "US"')

	Date	Province/State	Country	Lat	Long	Confirmed	Recovered	Deaths	Active
189465	2020-01-22		US	40.0	-100.0	1	0	0	1
189466	2020-01-23		US	40.0	-100.0	1	0	0	1
189467	2020-01-24		US	40.0	-100.0	2	0	0	2
189468	2020-01-25		US	40.0	-100.0	2	0	0	2
189469	2020-01-26		US	40.0	-100.0	5	0	0	5
...	...	...	...	...	...	...	...	...	...
190203	2022-01-29		US	40.0	-100.0	74232238	0	884405	73347833
190204	2022-01-30		US	40.0	-100.0	74424305	0	884726	73539579
190205	2022-01-31		US	40.0	-100.0	74951445	0	887148	74064297
190206	2022-02-01		US	40.0	-100.0	75350359	0	890770	74459589
190207	2022-02-02		US	40.0	-100.0	75680487	0	894316	74786171

743 rows × 9 columns

Scatterplot Visualization

fig = go.Figure()
fig.add_trace(go.Scatter(x = confirmed['Date'], y= confirmed['Confirmed'], mode='lines+markers', name='Confirmed Cases', line=dict(color='Orange',width =2)))
fig.add_trace(go.Scatter(x = recovered['Date'], y= recovered['Recovered'], mode='lines+markers', name='Recovered Cases', line=dict(color='Green', width =2)))
fig.add_trace(go.Scatter(x = deaths['Date'], y= deaths['Deaths'], mode='lines+markers', name='Deaths', line=dict(color='Red', width=2)))

fig.update_layout(title = "WorldWide Covid-19 Cases", xaxis_tickfont_size=14, yaxis= dict(title='Number of Cases'))
fig.show()

Cases Density Animation On World Map

#convert date to string
df['Date'] = df['Date'].astype(str)

df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 212498 entries, 0 to 212497
Data columns (total 9 columns):
 #   Column          Non-Null Count   Dtype  
---  ------          --------------   -----  
 0   Date            212498 non-null  object 
 1   Province/State  212498 non-null  object 
 2   Country         212498 non-null  object 
 3   Lat             212498 non-null  float64
 4   Long            212498 non-null  float64
 5   Confirmed       212498 non-null  int64  
 6   Recovered       212498 non-null  int64  
 7   Deaths          212498 non-null  int64  
 8   Active          212498 non-null  int64  
dtypes: float64(2), int64(4), object(3)
memory usage: 14.6+ MB

#using plotly express to display a density map
fig =px.density_mapbox(df, lat='Lat', lon='Long', hover_name='Country', hover_data=['Confirmed', 'Recovered', 'Deaths'], animation_frame='Date',color_continuous_scale='Portland', radius=7, zoom=0,height=700)
fig.update_layout(title='WorldWide Covid-19 Cases with TimeLapse')
fig.update_layout(mapbox_style='open-street-map', mapbox_center_lon=0)
fig.show()

Cases over a time period with Area Plot

df['Date'] =pd.to_datetime(df['Date'])
df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 212498 entries, 0 to 212497
Data columns (total 9 columns):
 #   Column          Non-Null Count   Dtype         
---  ------          --------------   -----         
 0   Date            212498 non-null  datetime64[ns]
 1   Province/State  212498 non-null  object        
 2   Country         212498 non-null  object        
 3   Lat             212498 non-null  float64       
 4   Long            212498 non-null  float64       
 5   Confirmed       212498 non-null  int64         
 6   Recovered       212498 non-null  int64         
 7   Deaths          212498 non-null  int64         
 8   Active          212498 non-null  int64         
dtypes: datetime64[ns](1), float64(2), int64(4), object(2)
memory usage: 14.6+ MB

temp = df.groupby('Date')['Confirmed', 'Deaths', 'Recovered', 'Active'].sum().reset_index()
temp = temp[temp['Date']==max(temp['Date'])].reset_index(drop =True)
tm =temp.melt(id_vars = 'Date', value_vars = ['Active', 'Deaths', 'Recovered'])
fig = px.treemap(tm , path = ['variable'], values='value', height = 250, width = 800, color_discrete_sequence=[act, dth, rec])
fig.data[0].textinfo ='label+text+value'
fig.show()

temp = df.groupby('Date')['Recovered', 'Deaths', 'Active'].sum().reset_index()
temp = temp.melt(id_vars = 'Date', value_vars =['Recovered', 'Deaths', 'Active'], var_name = 'Case', value_name = 'Count')

fig = px.area(temp, x='Date', y='Count', color='Case', height=500, title='Cases Over Time', color_discrete_sequence=[rec, dth, act])
fig.update_layout(xaxis_rangeslider_visible=True)
fig.show()

Cases visualization using Folium Maps

#Worldwide cases on Folium Maps
temp =df[df['Date'] == max(df['Date'])]

m =folium.Map(location =[0,0], tiles='cartodbpositron', min_zoom =1, max_zoom= 4, zoom_start=1)

for i in range(0, len(temp)):
    folium.Circle(location = [temp.iloc[i]['Lat'], temp.iloc[i]['Long']], color='crimson', fill =  'crimson', 
                tooltip = '<li><bold> Country: ' + str(temp.iloc[i]['Country'])+
                          '<li><bold> Province: ' + str(temp.iloc[i]['Province/State'])+
                          '<li><bold> Confirmed: ' + str(temp.iloc[i]['Confirmed'])+
                          '<li><bold> Deaths: ' + str(temp.iloc[i]['Deaths']),
                radius = int(temp.iloc[i]['Confirmed'])**0.5).add_to(m)

m

Make this Notebook Trusted to load map: File -> Trust Notebook

Confirmed Cases with Choropleth Map

fig = px.choropleth(country_daywise, locations='Country', locationmode='country names', color=country_daywise['Confirmed'],
                   hover_name = 'Country', animation_frame=country_daywise['Date'].dt.strftime('%Y-%m-%d'),
                   title = 'Cases Over Time in Different Countries', color_continuous_scale=px.colors.sequential.Inferno)

fig.update(layout_coloraxis_showscale=True)
fig.show()

Confirmed & Death Cases using Bar plots

fig_c =px.bar(daywise, x='Date', y='Confirmed', color_discrete_sequence=[act])
fig_d =px.bar(daywise, x='Date', y='Deaths', color_discrete_sequence=[dth])

fig = make_subplots(rows =1, cols =2, shared_xaxes=False, horizontal_spacing= 0.1,
                   subplot_titles=('Confirmed Cases', 'Death Cases'))

fig.add_trace(fig_c['data'][0], row=1, col=1)
fig.add_trace(fig_d['data'][0], row=1, col=2)

fig.update_layout(height=400)
fig.show()

Confirmed & Death Cases with Static Colormap

fig_c = px.choropleth(countrywise, locations='Country', locationmode='country names',
                     color=np.log(countrywise['Confirmed']), hover_name='Country', hover_data=['Confirmed'])

temp = countrywise[countrywise['Deaths']> 0]
fig_d = px.choropleth(temp, locations='Country', locationmode='country names',
                     color=np.log(temp['Deaths']), hover_name='Country', hover_data=['Deaths'])

fig = make_subplots(rows=1, cols=2, subplot_titles=['Confirmed Cases', 'Deaths'], 
                   specs=[[{'type':'choropleth'},{'type':'choropleth'}]])

fig.add_trace(fig_c['data'][0], row=1, col=1)
fig.add_trace(fig_d['data'][0], row=1, col=2)

fig.update(layout_coloraxis_showscale=False)

fig.show()

Deaths & Recoveries per 100 cases

Note: The Data is only accurate till July 2021, as after that time the data for recovery cases were not reported or were not being tracked by the institutes. And it is difficult to predict the number based on it.

daywise.columns

Index(['Date', 'Confirmed', 'Deaths', 'Recovered', 'Active', 'New Cases',
       'Deaths / 100 Cases', 'Recovered / 100 Cases', 'Deaths / 100 Recovered',
       'No. of Countries'],
      dtype='object')

fig1 = px.line(daywise, x='Date', y='Deaths / 100 Cases', color_discrete_sequence=[dth])
fig2 = px.line(daywise, x='Date', y='Recovered / 100 Cases', color_discrete_sequence=[rec])
fig3 = px.line(daywise, x='Date', y='Deaths / 100 Recovered', color_discrete_sequence=[rec])

fig = make_subplots(rows =1, cols=3, shared_xaxes=False, 
                   subplot_titles=('Deaths / 100 Cases','Recovered / 100 Cases','Deaths / 100 Recovered'))

fig.add_trace(fig1['data'][0], row=1,col=1)
fig.add_trace(fig2['data'][0], row=1,col=2)
fig.add_trace(fig3['data'][0], row=1,col=3)

fig.update_layout(height=400)
fig.show()

Number of new cases per Day & countries affected

fig_c = px.bar(daywise, x='Date', y='Confirmed', color_discrete_sequence=[act])
fig_d = px.bar(daywise, x='Date', y='No. of Countries', color_discrete_sequence=[dth])

fig =make_subplots(rows =1, cols=2, shared_xaxes=False, horizontal_spacing=0.1, subplot_titles=('No. of New Cases per Day', 'No. of Countries'))

fig.add_trace(fig_c['data'][0], row=1, col=1)
fig.add_trace(fig_d['data'][0], row=1, col=2)

fig.show()

Top 15 Countries Case Analysis

countrywise.columns

Index(['Country', 'Confirmed', 'Deaths', 'Recovered', 'Active', 'New Cases',
       'Deaths / 100 Cases', 'Recovered / 100 Cases', 'Deaths / 100 Recovered',
       'Population', 'Cases / Million People', 'Confirmed last week',
       '1 week change', '1 week % increase'],
      dtype='object')

%%javascript
IPython.OutputArea.prototype._should_scroll = function(lines) {
    return false;
}

top =15

#fig for confirmed cases
fig_c = px.bar(countrywise.sort_values('Confirmed').tail(top), x='Confirmed', y='Country',
              text = 'Confirmed', orientation='h', color_discrete_sequence=[cnf])

#fig for death cases
fig_d = px.bar(countrywise.sort_values('Deaths').tail(top), x='Deaths', y='Country',
              text = 'Deaths', orientation='h', color_discrete_sequence=[dth])

#fig for active cases
fig_a = px.bar(countrywise.sort_values('Active').tail(top), x='Active', y='Country',
              text = 'Active', orientation='h', color_discrete_sequence=[act])

#for recovered cases but due to discrepancy and no data available commented this part
# fig_r = px.bar(countrywise.sort_values('Recovered').tail(top), x='Recovered', y='Country',
#               text = 'Recovered', orientation='h', color_discrete_sequence=[rec])

#note we haven't plotted the fig's for any scenario where recovered cases is being used to calculate the result to avoid irregularity

#fig for deaths / 100 cases
fig_dc = px.bar(countrywise.sort_values('Deaths / 100 Cases').tail(top), x='Deaths / 100 Cases', y='Country',
              text = 'Deaths / 100 Cases', orientation='h', color_discrete_sequence=['#ff0000'])


#fig for new cases country wise
fig_nc = px.bar(countrywise.sort_values('New Cases').tail(top), x='New Cases', y='Country',
              text = 'New Cases', orientation='h', color_discrete_sequence=['#944dff'])

temp= countrywise[countrywise['Population']>1000000]
#fig for cases per million people
fig_p = px.bar(temp.sort_values('Cases / Million People').tail(top), x='Cases / Million People', y='Country',
              text = 'Cases / Million People', orientation='h', color_discrete_sequence=['#3366ff'])

#fig for 1 week changes
fig_ow = px.bar(countrywise.sort_values('1 week change').tail(top), x='1 week change', y='Country',
              text = '1 week change', orientation='h', color_discrete_sequence=['#ff6600'])

#fig for 1 week % increase
tem= countrywise[countrywise['Confirmed']>100]
fig_op = px.bar(tem.sort_values('1 week % increase').tail(top), x='1 week % increase', y='Country',
              text = '1 week % increase', orientation='h', color_discrete_sequence=['#990033'])



fig= make_subplots(rows=4 , cols=2, shared_xaxes=False, horizontal_spacing= 0.2, vertical_spacing=.05,
                  subplot_titles=('Confirmed Cases', 'Deaths Reported', 'Active Cases', 'Deaths / 100 Cases'
                                 ,'New Cases', 'Cases / Million People', '1 week change', '1 week % increase'))

fig.add_trace(fig_c['data'][0], row=1, col=1)
fig.add_trace(fig_d['data'][0], row=1, col=2)

fig.add_trace(fig_a['data'][0], row=2, col=1)
fig.add_trace(fig_dc['data'][0], row=2, col=2)


fig.add_trace(fig_nc['data'][0], row=3, col=1)
fig.add_trace(fig_p['data'][0], row=3, col=2)


fig.add_trace(fig_ow['data'][0], row=4, col=1)
fig.add_trace(fig_op['data'][0], row=4, col=2)


fig.update_layout(height=3000)
fig.show()

Scatter Plot for Deaths Vs Confirmed Cases

countrywise.sort_values('Deaths', ascending=False).head(15)

	Country	Confirmed	Deaths	Recovered	Active	New Cases	Deaths / 100 Cases	Recovered / 100 Cases	Deaths / 100 Recovered	Population	Cases / Million People	Confirmed last week	1 week change	1 week % increase
182	US	74951445	887148	0	74064297	527140	1.18	0.0	0.0	330541757	226753.0	71783483	3167962	4.41
23	Brazil	25463530	627589	0	24835941	102883	2.46	0.0	0.0	422706534	60239.0	24142032	1321498	5.47
79	India	41469499	496242	0	40973257	167059	1.20	0.0	0.0	2751341248	15072.0	39799202	1670297	4.20
144	Russia	11670366	324672	0	11345694	123033	2.78	0.0	0.0	292579178	39888.0	10988027	682339	6.21
115	Mexico	4942590	306091	0	4636499	12521	6.19	0.0	0.0	255584572	19338.0	4685767	256823	5.48
138	Peru	3224406	205505	0	3018901	0	6.37	0.0	0.0	65597846	49154.0	2976260	248146	8.34
186	United Kingdom	17431225	156281	0	17274944	848962	0.90	0.0	0.0	134862019	129252.0	16063010	1368215	8.52
85	Italy	10983116	146498	0	10836618	57631	1.33	0.0	0.0	120822834	90903.0	10001344	981772	9.82
80	Indonesia	4353370	144320	0	4209050	10185	3.32	0.0	0.0	273523621	15916.0	4289305	64065	1.49
37	Colombia	5887261	134300	0	5752961	15284	2.28	0.0	0.0	99141018	59383.0	5761398	125863	2.18
81	Iran	6373174	132454	0	6240720	28995	2.08	0.0	0.0	83992953	75877.0	6258181	114993	1.84
62	France	19266496	131937	0	19134559	88457	0.68	0.0	0.0	68128061	282798.0	16917220	2349276	13.89
6	Argentina	8378656	121273	0	8257383	43472	1.45	0.0	0.0	45195777	185386.0	7940657	437999	5.52
66	Germany	10025463	117979	0	9907484	179431	1.18	0.0	0.0	166310062	60282.0	8909503	1115960	12.53
184	Ukraine	4255206	106880	0	4148326	23063	2.51	0.0	0.0	89634808	47473.0	4055643	199563	4.92

top =15
fig = px.scatter(countrywise.sort_values('Deaths', ascending=False).head(top), 
                x = 'Confirmed', y='Deaths', color='Country', height=600, size='Confirmed',
                text='Country', log_x = True, log_y = True, title='Deaths vs Confirmed Cases(Cases are on log10 scale)')

fig.update_traces(textposition= 'top center')
fig.update_layout(showlegend = True)
fig.update_layout(xaxis_rangeslider_visible = True)
fig.show()

Stacked Bar Plot

fig = px.bar(country_daywise, x='Date', y='Confirmed', color='Country', height=600,
            title='Confirmed Cases', color_discrete_sequence=px.colors.cyclical.mygbm)
fig.show()

fig = px.bar(country_daywise, x='Date', y='Deaths', color='Country', height=600,
            title='Deaths', color_discrete_sequence=px.colors.cyclical.mygbm)
fig.show()

fig = px.bar(country_daywise, x='Date', y='New Cases', color='Country', height=600,
            title='New Cases', color_discrete_sequence=px.colors.cyclical.mygbm)
fig.show()

Line Plot

fig = px.line(country_daywise, x ='Date', y='Confirmed', color='Country', height=600, title='Confirmed Cases',
             color_discrete_sequence=px.colors.cyclical.mygbm)
fig.show()

fig = px.line(country_daywise, x ='Date', y='Deaths', color='Country', height=600, title='Deaths',
             color_discrete_sequence=px.colors.cyclical.mygbm)
fig.show()

fig = px.line(country_daywise, x ='Date', y='New Cases', color='Country', height=600, title='New Cases',
             color_discrete_sequence=px.colors.cyclical.mygbm)
fig.show()

Growth rate after 100 Cases

df['Date'] =pd.to_datetime(df['Date'])
df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 212498 entries, 0 to 212497
Data columns (total 9 columns):
 #   Column          Non-Null Count   Dtype         
---  ------          --------------   -----         
 0   Date            212498 non-null  datetime64[ns]
 1   Province/State  212498 non-null  object        
 2   Country         212498 non-null  object        
 3   Lat             212498 non-null  float64       
 4   Long            212498 non-null  float64       
 5   Confirmed       212498 non-null  int64         
 6   Recovered       212498 non-null  int64         
 7   Deaths          212498 non-null  int64         
 8   Active          212498 non-null  int64         
dtypes: datetime64[ns](1), float64(2), int64(4), object(2)
memory usage: 14.6+ MB

gt_100 = country_daywise[country_daywise['Confirmed']>100]['Country'].unique()

temp = df[df['Country'].isin(gt_100)]

temp = temp.groupby(['Country', 'Date'])['Confirmed'].sum().reset_index()
temp =temp[temp['Confirmed']>100]

min_date = temp.groupby('Country')['Date'].min().reset_index()
min_date.columns = ['Country', 'Min Date']

from_100th_case = pd.merge(temp, min_date, on ='Country')
from_100th_case['N days'] = (from_100th_case['Date'] - from_100th_case['Min Date']).dt.days

fig = px.line(from_100th_case, x = 'N days', y='Confirmed', color='Country', title='N Days From 100 Cases', height=600)
fig.show()

Growth rate after 1000 Cases

gt_1000 = country_daywise[country_daywise['Confirmed']>1000]['Country'].unique()
temp = df[df['Country'].isin(gt_1000)]

temp = temp.groupby(['Country', 'Date'])['Confirmed'].sum().reset_index()
temp =temp[temp['Confirmed']>1000]

min_date = temp.groupby('Country')['Date'].min().reset_index()
min_date.columns = ['Country', 'Min Date']

from_1000th_case = pd.merge(temp, min_date, on ='Country')
from_1000th_case['N days'] = (from_1000th_case['Date'] - from_1000th_case['Min Date']).dt.days

fig = px.line(from_1000th_case, x = 'N days', y='Confirmed', color='Country', title='N Days From 1000 Cases', height=600)
fig.show()

Growth rate after 100K Cases

gt_100000 = country_daywise[country_daywise['Confirmed']>100000]['Country'].unique()
temp = df[df['Country'].isin(gt_100000)]

temp = temp.groupby(['Country', 'Date'])['Confirmed'].sum().reset_index()
temp =temp[temp['Confirmed']>100000]

min_date = temp.groupby('Country')['Date'].min().reset_index()
min_date.columns = ['Country', 'Min Date']

from_100000th_case = pd.merge(temp, min_date,on ='Country')
from_100000th_case['N days'] = (from_100000th_case['Date'] - from_100000th_case['Min Date']).dt.days

fig = px.line(from_100000th_case, x = 'N days', y='Confirmed', color='Country', title='N Days From 100K Cases', height=600)
fig.show()

Growth rate after 1Million Cases

gt_1000000 = country_daywise[country_daywise['Confirmed']>1000000]['Country'].unique()
temp = df[df['Country'].isin(gt_1000000)]

temp = temp.groupby(['Country', 'Date'])['Confirmed'].sum().reset_index()
temp =temp[temp['Confirmed']>1000000]

min_date = temp.groupby('Country')['Date'].min().reset_index()
min_date.columns = ['Country', 'Min Date']

from_1000000th_case = pd.merge(temp, min_date,on ='Country')
from_1000000th_case['N days'] = (from_1000000th_case['Date'] - from_1000000th_case['Min Date']).dt.days

fig = px.line(from_1000000th_case, x = 'N days', y='Confirmed', color='Country', title='N Days From 1Million Cases', height=600)
fig.show()

Growth rate after 10Million Cases

gt_10000000 = country_daywise[country_daywise['Confirmed']>10000000]['Country'].unique()
temp = df[df['Country'].isin(gt_10000000)]

temp = temp.groupby(['Country', 'Date'])['Confirmed'].sum().reset_index()
temp =temp[temp['Confirmed']>10000000]

min_date = temp.groupby('Country')['Date'].min().reset_index()
min_date.columns = ['Country', 'Min Date']

from_10000000th_case = pd.merge(temp, min_date,on ='Country')
from_10000000th_case['N days'] = (from_10000000th_case['Date'] - from_10000000th_case['Min Date']).dt.days

fig = px.line(from_10000000th_case, x = 'N days', y='Confirmed', color='Country', title='N Days From 10Million Cases', height=600)
fig.show()

Tree Map Analysis

# Confirmed Cases
full_latest = df[df['Date'] == max(df['Date'])]

fig = px.treemap(full_latest.sort_values(by='Confirmed', ascending=False).reset_index(drop=True),
                path = ['Country', 'Province/State'], values='Confirmed', height=700,
                title='Number of Confirmed Cases',
                color_discrete_sequence=px.colors.qualitative.Dark2)

fig.data[0].textinfo= 'label+text+value'
fig.show()

#deaths
full_latest = df[df['Date'] == max(df['Date'])]

fig = px.treemap(full_latest.sort_values(by='Deaths', ascending=False).reset_index(drop=True),
                path = ['Country', 'Province/State'], values='Deaths', height=700,
                title='Number of Deaths',
                color_discrete_sequence=px.colors.qualitative.Dark2)

fig.data[0].textinfo= 'label+text+value'
fig.show()

First & Last Case Report Time

first_date= df[df['Confirmed']>0]
first_date = first_date.groupby('Country')['Date'].agg(['min']).reset_index()


last_date= df.groupby(['Country', 'Date'])['Confirmed', 'Deaths']
last_date = last_date.sum().diff().reset_index()


mask = (last_date['Country'] != last_date['Country'].shift(1))

last_date.loc[mask, 'Confirmed'] = np.nan
last_date.loc[mask, 'Deaths'] = np.nan

last_date = last_date[last_date['Confirmed']>0]
last_date = last_date.groupby('Country')['Date'].agg(['max']).reset_index()

first_last = pd.concat([first_date, last_date['max']], axis=1)
first_last['max'] = first_last['max'] + timedelta(days=1)

first_last['Days'] = first_last['max'] - first_last['min']
first_last['Task'] = first_last['Country'] 

first_last.columns = ['Country', 'Start', 'Finish', 'Days', 'Task']

first_last = first_last.sort_values('Days')

colors = ['#' + ''.join([random.choice('0123456789ABCDEF') for j in range(6)]) for i in range(len(first_last))]


fig = ff.create_gantt(first_last, index_col ='Country', colors=colors, show_colorbar=False,
                     bar_width =0.2, showgrid_x = True, showgrid_y = True, height= 2500)
fig.show()

Confirmed Cases Country & Day Wise

temp = country_daywise.groupby(['Country', 'Date'])['Confirmed'].sum().reset_index()
temp = temp[temp['Country'].isin(gt_10000000)]

countries= temp['Country'].unique()


ncols = 3
nrows = math.ceil(len(countries)/ncols)

fig = make_subplots(rows = nrows, cols= ncols, shared_xaxes= False, subplot_titles= countries)

for ind, country in enumerate(countries):
    row = int((ind/ncols)+1)
    col = int((ind%ncols)+1)
    fig.add_trace(go.Bar(x= temp['Date'], y=temp.loc[temp['Country']== country, 'Confirmed'], name=country), row= row, col=col)
    
fig.update_layout(height=4000, title_text='Confirmed Cases in each Country')    
fig.update_layout(showlegend=False)
fig.show()

Covid-19 Vs Other Similar Epidemics

#source for the number wikipedia
epidemics = pd.DataFrame({
    'epidemic' :['COVID-19', 'SARS', 'EBOLA', 'MERS', 'H1N1'],
    'start_year' : [2019, 2002, 2013, 2012, 2009],
    'end_year' :[2020, 2004, 2016, 2020, 2010],
    'confirmed' :[full_latest['Confirmed'].sum(), 8422, 28646, 2519, 6724149],
    'deaths': [full_latest['Deaths'].sum(), 813, 11323, 866, 19654]
})

#calculating mortality rate
epidemics['mortality'] =round((epidemics['deaths']/epidemics['confirmed'])*100, 2)

temp = epidemics.melt(id_vars='epidemic', value_vars=['confirmed', 'deaths', 'mortality'], 
                     var_name='Case', value_name='Value')

fig= px.bar(temp, x='epidemic', y='Value', color= 'epidemic', text='Value', facet_col='Case',
           color_discrete_sequence=px.colors.qualitative.Bold)

fig.update_traces(textposition='outside')
fig.update_layout(uniformtext_minsize=8, uniformtext_mode= 'hide')
fig.update_yaxes(showticklabels= False)
fig.layout.yaxis2.update(matches = None)
fig.layout.yaxis3.update(matches = None)
fig.show()